Ionizing radiation kills cells and causes genome rearrangements primarily by introducing double strand breaks (DSBs) in chromosomes. These chromosome breaks are repaired by homologous recombination or the less accurate nonhomologous end joining pathway (NHEJ). NHEJ is also the only pathway for efficient repair of DSB intermediates in V(D)J recombination. Deficiency in NHEJ thus leads to radiosensitivity, severe immunodeficiency, genetic instability, and cancer predisposition. 1) NHEJ employs three DNA polymerases: pol lambda, pol mu, and TdT. Prior work hints that different substrate requirements have a critical impact on their biological role. We will now systematically compare activities of these three polymerases on different substrates, to determine what advantage there is to having all three work in the same repair pathway. 2) The structural basis for how and why three related, but different polymerases are specifically associated with NHEJ will be investigated. In collaboration with two structural biology labs we will generate mutant and chimeric versions of the three polymerases. The impact of these changes on polymerase activity during NHEJ will then be assessed using sophisticated cell-free and cellular assays. 3) Homologous recombination, though more accurate than NHEJ, is generally more demanding of DNA synthesis. However, levels of synthesis precursors fluctuate greatly during the cell cycle and in different cell types. We will manipulate levels of synthesis precursors, and determine if this is a major criteria for choosing a particular double strand break repair pathway. NHEJ is less accurate than most repair pathways, yet it still suppresses tumorigenesis. This research will help reconcile these observations: in other words, we will determine why NHEJ might be better than the alternative. The ability to manipulate the choice of cellular repair pathway by depletion of DNA synthesis precursors, as suggested in our third aim, could also improve treatment of tumors by radiation. Repair of chromosome breaks by nonhomologous end joining (NHEJ) is less accurate than most repair pathways, yet it still suppresses tumorigenesis. This research will help reconcile these observations: in other words, we will determine why NHEJ might be better than the alternative. Additionally, a facile ability to manipulate the choice of cellular repair pathway, as suggested by preliminary work, could lead to improved treatment of tumors by radiation.